1 In this connection, the reliabilist Goldman, Alvin I. writes: “…general reliability is probably best understood as a propensity rather than a frequency. This avoids the possibility that actual uses of the process are numerically too limited or skewed torepresentan intuitively appropriate ratio” (Epistemology and Cognition [Cambridge, MA: Harvard University Press, 1986], p. 49)Google Scholar. By a “propensity” of anything (such as a belief-forming process) Goldman evidently means here a sort of irreducibly modal fact about what it would tend, under suitable conditions, to do, that is, something about it that can in no way be reduced to facts about what it or anything else has done, is doing and will do actually. But o t postulating facts of that sort, nominalistically inclined philosophers stand-ardly take strong exception. At any rate, in my paper, “The Truth-conditions of Counterfactual Conditional Sentences” (Mind, 87, 345 [01 1978]: 1–21)Google Scholar, I advanced a construal of subjunctive-conditional constructions making no reference at all topossible worlds, Platonic universals, or irreducibly modal propen-sities—though the account given does imply the dependence of many subjunctive conditionals' truth upon objective principles of inductive probabil-ification. The hope is that this approach to subjunctive conditionals can secure all of the advantages Goldman seeks from “propensities” without incurring the obvious ontological objections to which talk of them gives rise. The paper cited seeks also to explain the conditions for the truth of subjunctive conditional sentences whose protases introduce suppositions that are not only contrary-to-fact but contrary-to-law, or even logically contradictory. On that basis, the belief in a logical necessity such as the fact that 7 + 5 = 12 can meaningfully be described as due to causes that would not have operated to produce the belief if it were not that 7 + 5 = 12.

2 Cf. Griffiths, A. Phillips, “Introduction,” in Knowledge and Belief, edited by Griffiths, A. P. (Oxford: Oxford University Press, 1967), p. 12Google Scholar; Bonjour, Laurence, “Externalist Theories of Empirical Knowledge,” in Midwest Studies in Philosophy V: Studies in Epistemology, edited by French, Peter A., Uehling, Theodore E. Jr, and Wettstein, Howard K. (Minneapolis, MN: University of Minnesota Press, 1980), pp. 59–62;Google ScholarBonjour, Laurence, The Structure of Empirical Knowledge (Cambridge, MA: Harvard University Press, 1985), p. 38–40; and myGoogle Scholar“A Contribution Towards the Development of the Causal Theory of Knowledge,” Australasian Journalof Philosophy, 50, 3 (12 1972), pp. 247–48. However, as will be seen below, I ought to have stipulated in that paper that a true belief does not constitute knowledge in the case where the process producing and sustaining it would not operate but for some false opinion on the believer's part that forms part of the believer's grounds for holding the beliefGoogle Scholar.

3 Bonjour, , The Structure of Empirical Knowledge, p. 42Google Scholar.

4 See my “Causal Theory of Knowledge,” pp. 244–45.

5 See my paper, “Immorality with a Clear Conscience,” American Philosophical Quarterly, 17, 3 (07 1980): 245–50Google Scholar.

6 Goldstick, , “Cognitive Reason,” Philosophy and Phenomenological Research, 52, 1 (03 1992): 117–24CrossRefGoogle Scholar.

7 Could we regard the process of coming to accordbeliefto truths while with holding it from falsehoods as a type of causal process of the sort in question here—a type of process exemplified, perhaps, in all true belief inceptions? No, we could not, i f any more specific description of a type of process than this is required before what is described can count as a type of causal process. Just what about such a process does make it one that would generallyresultin true, and not false, belief? Where that question has an answer, a causal type of process has not been specified merely by the description “process of coming to accord belief to truths while with holding it from falsehoods.” Could there exist a causality whereby a mind was caused to accord belief to truths while withholding it from falsehoods without there being any answer to the question what it specifically was about that causality which made it have just such aresult? Something of this sort is certainly attributed by many theists to the cognitive life of the Divine mind. For our purposes there is no need to disallow such a causality by definition. But, where an answer exists (known or unknown) to the question what it specifically is about a type of belief-forming process that makes it one that would be generally reliable, we can insist that no causal process type has been described unless the description offered actually does expressly mention the feature(s) in question (viz., the specific one[s] responsible for this type-process's causal tendency to yield just that result).

8 Ginet, Carl, “Contra Reliabilism,” The Monist, 68, 2 (04 1985), quoted from p. 176CrossRefGoogle Scholar.

9 Ibid., p. 175.

10 See my “Counterfactual Conditional Sentences.”

11 Pollock, John L., “Reliability and Justified Belief,” Canadian Journal of Philosophy, 14, 1 (03 1984), see p. 104CrossRefGoogle Scholar.

12 Ibid., pp. 105–106.

13 Ibid.

14 Ibid., p. 106.

15 paper, Jacobson's, “Why Epistemic Reasons Need Not Be Causes,” was read to a session of the Canadian Philosophical Association in Hamilton, Ontario, on 05 24, 1987. (J. C. Poggendorff's Biographisch-Literarisches Handwörterbuch zur Geschichte der Exacten Wissenschaften does not list any nineteenth-century mathematician named Sussfonder. Could we, then, just translate “Sussföender” as “blow-dried sweetie”?)Google Scholar

16 These two paragraphs are largely reproduced from my formally pre-scheduled comments on Murray Clarke's paper, “The Background of Justification,” which were read to a session of the Canadian Philosophical Association in Winnipeg on May 27,1986.